Annually in the U.S., about 250,000 women are diagnosed with breast cancer, and 40,000 die from it. The majority (~75%) of cases appear to have a sporadic, rather than an inherited basis. This has concentrated efforts on identifying common spontaneous genomic alterations that may underlie carcinogenesis. This approach is complicated by the genetic heterogeneity of patients and variety of tumor types. Mouse models are powerful for untangling this problem because such heterogeneities can be controlled. In previous studies, we found that mice bearing the genomic instability mutation Chaos3, an allele of the replicative helicase component Mcm4, caused exclusively mammary adenocarcinomas in nearly all homozygous nulliparous females. Strikingly, nearly all Chaos3 mammary tumors underwent deletions of Nf1 (Neurofibromin 1). NF1 is a tumor suppressor that negatively regulates the Ras oncogene. Recent genomic studies suggest that spontaneous mutation of NF1 contributes to several human cancers, though a potential role in breast cancer hasn't been established. Remarkably, our examination of human breast cancer genome datasets revealed that >27% of human breast tumors have deletions (primarily) or mutations of NF1. Combined with the recurrent loss of Nf1 in Chaos3 mouse mammary tumors, these data implicate NF1 as an important breast tumor suppressor. We propose to use mouse models to test directly whether inherited or induced Nf1 deficiency contributes to mammary tumors. Additionally, using Chaos3 mammary and human breast cancer cell lines for tumor reconstitution experiments in mice, we will determine if NF1 loss is needed for maintenance of NF1-mutated cancers. In sum, this exploratory R21 project will rigorously test the hypothesis that NF1 is a significant breast cancer susceptibility gene. Validation of this hypothesis would impact the treatment of patients with NF1-mutated breast cancers, and provide a powerful model for therapeutic development.